Alhassane Diallo and colleagues conducted an interesting and meaningful meta-analysis.1 In the study, authors assessed the subgroup effects of glucagon-like peptide 1 receptor agonists (GLP-1RAs) and sodium-glucose cotransporter-2 inhibitors (SGLT2i) on major adverse cardiovascular event (MACE) by sex, age, body mass index, race, and duration of diabetes. Accordingly, they concluded: a greater benefit of GLP-1RAs on MACE was found in elderly patients and in Asian patients, and that of SGLT2i in patients with a long diabetes duration. However, the reliability of these findings might be adversely influenced by the following issues.
First, the multiple testing issue seemed not to be correctly considered. In the article, authors in total calculated 12 important P values for subgroup differences (Pinteraction): Figs. 2–4 report 4 Pinteraction, respectively. According to the Bonferroni method,2 the corrected significance threshold should be calculated: 0.05 ÷ 12 = 0.0042. According to this threshold, three subgroup effects stated in the Conclusion (For SGLT2i: long-term versus short duration, Pinteraction = 0.03. For GLP-1RAs: Asian versus White, Pinteraction = 0.07; ≥75 versus <75 years, Pinteraction = 0.37) would be far from statistical significance. If Pinteraction calculated in Figs. S3 and S4 was additionally considered, the subgroup effects proposed by authors would be further from corrected significance. Since the Bonferroni correction frequently is too strict and sacrifices power,3 authors should identify a more appropriate method to address the multiple testing issue.
Second, Fig. 4 in a secondary analysis article4 of the CREDENCE trial reports all of the subgroup data of interest for this meta-analysis.1 However, authors did not include them.
Declaration of interests
All authors declare no competing interests.
Acknowledgments
This work was supported by the Key Laboratory of Neuroimaging, Longhua District, Shenzhen (Shen Long Hua Ke Chuang Ke Ji Zi (2022) No. 7); and Shenzhen Fundamental Research Program, General Programe for Fundamental Research (Grant No. JCYJ20210324142404012).
Contributor Information
Mei Qiu, Email: 13798214835@sina.cn.
Lixin Du, Email: dulixin976@yeah.net.
References
- 1.Diallo A., Carlos-Bolumbu M., Galtier F. Age, sex, race, BMI, and duration of diabetes differences in cardiovascular outcomes with glucose lowering drugs in type 2 diabetes: a systematic review and meta-analysis. eClinicalMedicine. 2022;54 doi: 10.1016/j.eclinm.2022.101697. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bland J.M., Altman D.G. Multiple significance tests: the Bonferroni method. BMJ. 1995;310:170. doi: 10.1136/bmj.310.6973.170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bender R., Lange S. Adjusting for multiple testing--when and how? J Clin Epidemiol. 2001;54:343–349. doi: 10.1016/s0895-4356(00)00314-0. [DOI] [PubMed] [Google Scholar]
- 4.Mahaffey K.W., Jardine M.J., Bompoint S., et al. Canagliflozin and cardiovascular and renal outcomes in type 2 diabetes mellitus and chronic kidney disease in primary and secondary cardiovascular prevention groups. Circulation. 2019;140:739–750. doi: 10.1161/CIRCULATIONAHA.119.042007. [DOI] [PMC free article] [PubMed] [Google Scholar]